• KDI Journal of Economic Policy
• /
• v.32 no.4
• /
• pp.115-151
• /
• 2010

Recently Korean universities show very rapid increases in both patents and R&D (research and development) expenditures. During the period from 1970 to 2008, university R&D spending has on the average increased 15.3% annually. Along with steady increases in R&D spending, university's research outputs have also continuously increased. In 1990 Korea as a total published 1,613 SCI-level scientific papers and Korean universities applied 27 patents to Korea patent office. In 2008, Korea published more that 35,000 SCI papers and Korean universities applied about 7,300 patents. The growth of scientific articles had begun from the early 1990s whereas the growth of patent has ignited entering the 2000s. The paper tried to investigate university research through the window of patent. Patents lie between invention and innovation and represent the potential value of invention which will be realized at the marketplace. Since Korean patents do not contain citation information, the paper used US patents-NBER patent database-as the main data. The key empirical question is whether Korean university patents granted from USPTO are characteristically different from other Korean patents granted from USPTO. Previous studies on US and Europe show that corporate patents are more stylized in appropriablity of invention, whereas university patents basicness. In case of Korea, the paper confirmed the appropriability characteristic of corporate patents; but the Korean unversity patents are not distinguishable in terms of basicness. The paper estimated the citation frequency function-an empirical model which was firstly developed by Caballero and Jaffe (1993) and later articulated by Jaffe and Trajtenberg (1996, 2002). The model is specified mainly composed of two interacting parts-diffusion effect and obsolescence effect of new ideas or innovations. Estimation results show that differences in forward citations between university and corporate patents are not statistically significant, after controlling self-citation. Since forward citations represent the quality of patents, this estimation result implies that there are no statistically significant quality differences between university and corporate patents. Prior research results, based on the same model of citation frequency function, about US and some European cases show that, in terms of forward citations, university patents are generally superior to corporate patents -for the case of US- or, the former not inferior to the latter-for the case of most of Europe. It is argued that some important and significant policy changes caused the rapid rise of university patents in Korea. Policy changes include the revision of technology transfer act allowing the ownership of publicly-funded research results to researchers and the changes in faculty/professor evaluation which gives more credit to the number of patents. These policy changes have triggered the rapid growth of the number of university patents. The results of the empirical analysis in this paper indicated that Korea now needs to make further efforts to enhance the quality of university patents, not just to produce more numbers of patents.

• Korean Journal of Food Science and Technology
• /
• v.32 no.3
• /
• pp.590-597
• /
• 2000

The rheological properties of ${\beta}-glucans$ isolated from non-waxy and waxy barley were investigated. ${\beta}-Glucan$ solutions showed pseudoplastic properties and their behaviors were explained by applying Power law model in the range of concentrations$(1{\sim}4%)$ and temperatures$(20{\sim}65^{\circ}C)$. The effects of temperature and concentration on the apparent viscosity at $700\;s^{-1}$ shear rate were examined by applying Arrhenius equation and power law equation, and their effect was more pronounced in waxy ${\beta}-glucan$ solutions. The activation energy for flow of ${\beta}-glucan$ solutions decreased with the increase of concentration, and the concentration-dependent constant A increased with the increase of temperature. The intrinsic viscosity of waxy ${\beta}-glucan$ was higher than that of non-waxy ${\beta}-glucan$. The transition from dilute to concentrate region occurred at a critical coil overlap parameter $C^*[{\eta}]=0.02.$ The slopes of non-waxy and waxy ${\beta}-glucan$ at $C[{\eta}] were similar, but the slope of waxy ${\beta}-glucan$ at $C[{\eta}]>C^*[{\eta}]$ was higher than that of non-waxy ${\beta}-glucan$. Dynamic viscoelasticity measurement showed that cross-over happened, and storage modulus was higher than loss modulus at frequency range above cross-over. ${\beta}-Glucan$ solutions formed weak gels after stored for 24 hr.

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.493-500
• /
• 2021

In this paper, we propose the development of deep learning structure to improve quality of polygonal containers. The deep learning structure consists of a convolution layer, a bottleneck layer, a fully connect layer, and a softmax layer. The convolution layer is a layer that obtains a feature image by performing a convolution 3x3 operation on the input image or the feature image of the previous layer with several feature filters. The bottleneck layer selects only the optimal features among the features on the feature image extracted through the convolution layer, reduces the channel to a convolution 1x1 ReLU, and performs a convolution 3x3 ReLU. The global average pooling operation performed after going through the bottleneck layer reduces the size of the feature image by selecting only the optimal features among the features of the feature image extracted through the convolution layer. The fully connect layer outputs the output data through 6 fully connect layers. The softmax layer multiplies and multiplies the value between the value of the input layer node and the target node to be calculated, and converts it into a value between 0 and 1 through an activation function. After the learning is completed, the recognition process classifies non-circular glass bottles by performing image acquisition using a camera, measuring position detection, and non-circular glass bottle classification using deep learning as in the learning process. In order to evaluate the performance of the deep learning structure to improve quality of polygonal containers, as a result of an experiment at an authorized testing institute, it was calculated to be at the same level as the world's highest level with 99% good/defective discrimination accuracy. Inspection time averaged 1.7 seconds, which was calculated within the operating time standards of production processes using non-circular machine vision systems. Therefore, the effectiveness of the performance of the deep learning structure to improve quality of polygonal containers proposed in this paper was proven.

• Information Systems Review
• /
• v.16 no.3
• /
• pp.161-177
• /
• 2014

Corporate credit rating assessment consists of complicated processes in which various factors describing a company are taken into consideration. Such assessment is known to be very expensive since domain experts should be employed to assess the ratings. As a result, the data-driven corporate credit rating prediction using statistical and artificial intelligence (AI) techniques has received considerable attention from researchers and practitioners. In particular, statistical methods such as multiple discriminant analysis (MDA) and multinomial logistic regression analysis (MLOGIT), and AI methods including case-based reasoning (CBR), artificial neural network (ANN), and multiclass support vector machine (MSVM) have been applied to corporate credit rating.2) Among them, MSVM has recently become popular because of its robustness and high prediction accuracy. In this study, we propose a novel optimized MSVM model, and appy it to corporate credit rating prediction in order to enhance the accuracy. Our model, named 'GAMSVM (Genetic Algorithm-optimized Multiclass Support Vector Machine),' is designed to simultaneously optimize the kernel parameters and the feature subset selection. Prior studies like Lorena and de Carvalho (2008), and Chatterjee (2013) show that proper kernel parameters may improve the performance of MSVMs. Also, the results from the studies such as Shieh and Yang (2008) and Chatterjee (2013) imply that appropriate feature selection may lead to higher prediction accuracy. Based on these prior studies, we propose to apply GAMSVM to corporate credit rating prediction. As a tool for optimizing the kernel parameters and the feature subset selection, we suggest genetic algorithm (GA). GA is known as an efficient and effective search method that attempts to simulate the biological evolution phenomenon. By applying genetic operations such as selection, crossover, and mutation, it is designed to gradually improve the search results. Especially, mutation operator prevents GA from falling into the local optima, thus we can find the globally optimal or near-optimal solution using it. GA has popularly been applied to search optimal parameters or feature subset selections of AI techniques including MSVM. With these reasons, we also adopt GA as an optimization tool. To empirically validate the usefulness of GAMSVM, we applied it to a real-world case of credit rating in Korea. Our application is in bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. The experimental dataset was collected from a large credit rating company in South Korea. It contained 39 financial ratios of 1,295 companies in the manufacturing industry, and their credit ratings. Using various statistical methods including the one-way ANOVA and the stepwise MDA, we selected 14 financial ratios as the candidate independent variables. The dependent variable, i.e. credit rating, was labeled as four classes: 1(A1); 2(A2); 3(A3); 4(B and C). 80 percent of total data for each class was used for training, and remaining 20 percent was used for validation. And, to overcome small sample size, we applied five-fold cross validation to our dataset. In order to examine the competitiveness of the proposed model, we also experimented several comparative models including MDA, MLOGIT, CBR, ANN and MSVM. In case of MSVM, we adopted One-Against-One (OAO) and DAGSVM (Directed Acyclic Graph SVM) approaches because they are known to be the most accurate approaches among various MSVM approaches. GAMSVM was implemented using LIBSVM-an open-source software, and Evolver 5.5-a commercial software enables GA. Other comparative models were experimented using various statistical and AI packages such as SPSS for Windows, Neuroshell, and Microsoft Excel VBA (Visual Basic for Applications). Experimental results showed that the proposed model-GAMSVM-outperformed all the competitive models. In addition, the model was found to use less independent variables, but to show higher accuracy. In our experiments, five variables such as X7 (total debt), X9 (sales per employee), X13 (years after founded), X15 (accumulated earning to total asset), and X39 (the index related to the cash flows from operating activity) were found to be the most important factors in predicting the corporate credit ratings. However, the values of the finally selected kernel parameters were found to be almost same among the data subsets. To examine whether the predictive performance of GAMSVM was significantly greater than those of other models, we used the McNemar test. As a result, we found that GAMSVM was better than MDA, MLOGIT, CBR, and ANN at the 1% significance level, and better than OAO and DAGSVM at the 5% significance level.

• Communications of Mathematical Education
• /
• v.24 no.1
• /
• pp.235-257
• /
• 2010

Contemporary belief is that the creative talented can create new knowledge and lead national development, so lots of countries in the world have interest in Gifted Education. As we well know, U.S.A., England, Russia, Germany, Australia, Israel, and Singapore enforce related laws in Gifted Education to offer Gifted Classes, and our government has also created an Improvement Act in January, 2000 and Enforcement Ordinance for Gifted Improvement Act was also announced in April, 2002. Through this initiation Gifted Education can be possible. Enforcement Ordinance was revised in October, 2008. The main purpose of this revision was to expand the opportunity of Gifted Education to students with special education needs. One of these programs is, the opportunity of Gifted Education to be offered to lots of the Gifted by establishing Special Classes at each school. Also, it is important that the quality of Gifted Education should be combined with the expansion of opportunity for the Gifted. Social opinion is that it will be reckless only to expand the opportunity for the Gifted Education, therefore, assessment on the Teaching and Learning Program for the Gifted is indispensible. In this study, 3 middle schools were selected for the Teaching and Learning Programs in mathematics. Each 1st Grade was reviewed and analyzed through comparative tables between Regular and Gifted Education Programs. Also reviewed was the content of what should be taught, and programs were evaluated on assessment standards which were revised and modified from the present teaching and learning programs in mathematics. Below, research issues were set up to assess the formation of content areas and appropriateness for Teaching and Learning Programs for the Gifted in mathematics. A. Is the formation of special class content areas complying with the 7th national curriculum? 1. Which content areas of regular curriculum is applied in this program? 2. Among Enrichment and Selection in Curriculum for the Gifted, which one is applied in this programs? 3. Are the content areas organized and performed properly? B. Are the Programs for the Gifted appropriate? 1. Are the Educational goals of the Programs aligned with that of Gifted Education in mathematics? 2. Does the content of each program reflect characteristics of mathematical Gifted students and express their mathematical talents? 3. Are Teaching and Learning models and methods diverse enough to express their talents? 4. Can the assessment on each program reflect the Learning goals and content, and enhance Gifted students' thinking ability? The conclusions are as follows: First, the best contents to be taught to the mathematical Gifted were found to be the Numeration, Arithmetic, Geometry, Measurement, Probability, Statistics, Letter and Expression. Also, Enrichment area and Selection area within the curriculum for the Gifted were offered in many ways so that their Giftedness could be fully enhanced. Second, the educational goals of Teaching and Learning Programs for the mathematical Gifted students were in accordance with the directions of mathematical education and philosophy. Also, it reflected that their research ability was successful in reaching the educational goals of improving creativity, thinking ability, problem-solving ability, all of which are required in the set curriculum. In order to accomplish the goals, visualization, symbolization, phasing and exploring strategies were used effectively. Many different of lecturing types, cooperative learning, discovery learning were applied to accomplish the Teaching and Learning model goals. For Teaching and Learning activities, various strategies and models were used to express the students' talents. These activities included experiments, exploration, application, estimation, guess, discussion (conjecture and refutation) reconsideration and so on. There were no mention to the students about evaluation and paper exams. While the program activities were being performed, educational goals and assessment methods were reflected, that is, products, performance assessment, and portfolio were mainly used rather than just paper assessment.